Clinical and epidemiological studies show a robust, inverse association of HDL cholesterol (HDL-C) levels with coronary heart disease (CHD) risk. However, raising HDL-C does not seem to provide therapeutic benefits, indicating that HDL-C levels are not a reliable measure of the cardio-protective effects of HDL. Moreover, the relationship between HDL-C and stroke risk is not clear with studies showing an inverse, direct, or lack of association. Recently, in mostly white cohorts, the cholesterol efflux capacity (CEC) of HDL correlated with both the prevalence and incidence of CHD independently of plasma levels of HDL-C and APOA1 (the major structural protein of HDL). On the other hand, increased HDL CEC was associated with increased stroke risk. Our preliminary data show no association between stroke risk and plasma HDL-C levels but a robust reduction in HDL CEC in stroke patients indicates that HDL's function might be a more relevant metric for assessing the stroke risk. Moreover, our preliminary results suggest that although African Americans have higher HDL-C levels they suffer from greater CHD risk, therefore, it is imperative to determine if HDL CEC associates with CHD risk better than HDL-C. We propose to determine if measures of HDL CEC do indeed provide value for CHD or stroke risk prediction in a racially balanced cohort. Studies in mostly white cohorts highlight that changes in HDL function are a reflection of changes in the HDL proteome. Therefore, identification of the protein cargo signature responsible for loss of HDL CEC in a bi- racial cohort may provide biomarkers of clinical validity that indicate CHD or stroke risk tailored to each race. To evaluate the predictive value of sterol efflux capacity of HDL and the protein signature that associates with dysfunctional HDL, we will use a case-control cohort (with 1000 each healthy, stroke and CHD cases) within The REasons for Geographic And Racial Differences in Stroke (REGARDS) Study. This is a NIH-sponsored national, longitudinal study of risk factors for stroke in over 30,000 African-American and Caucasian American men and women over 45 years of age, with a high prevalence of stroke and CHD. Due to unique composition of the REGARDS population we will be able to determine the impact of race and sex on our measures. Our preliminary data in mice using the 100-strain Hybrid Mouse Diversity Panel, developed by our collaborator, Dr. Jake Lusis of UCLA, successfully identified significant QTLs that modulate sterol efflux. It is therefore important to understand whether genetic mapping of sterol efflux trait identifies new CHD and stroke- related polymorphisms in humans. The data produced by this study would 1) determine the impact of race and sex on HDL function and the HDL proteome, 2) form the basis for development of clinical assays for assessing the role of HDL in cardiovascular disease, 3) facilitate race-specific prediction of CHD and stroke risk.